Liberating Minds


 
HomePortalCalendarFAQSearchRegisterMemberlistUsergroupsLog in
Share | .
 

 Anti-spam invention: How DOES it Work?

View previous topic View next topic Go down 
Go to page : 1, 2  Next
AuthorMessage
Zebra Foal



Number of posts: 899
Registration date: 2007-08-16

PostSubject: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 8:11 am

Can anyone explain to me how Steven Kirsch's method actually works, technically?--or know of any other explanatory links?

Would be unendingly grateful!


http://thehermesproject.blogspot.com/2007/12/99-effective-anti-spam-solutions.html

Monday, 3 December 2007
99% effective anti spam solutions?
The New York Times is profiling a very interesting approach to the problem of spam - profiling the recipient rather than the sender, and the creator is claiming a 99% success rate for his technology. And he even backs that up with a money back guarantee.

Serial Silicon Valley entrepreneur Steven Kirsch has invested USD$5m in his Abaca e-mail protection gateway solution, and is causing waves amongst the anti-spam industry.

Kirsch, who is suffering from a rare form of blood cancer (another problem he is intending to solve), said:


"We were sitting around thinking of ways to obfuscate the description about how our system worked so the spammers would be misdirected.

So I came up with receiver reputation as something that might sound plausible. Then as I thought about it more and more, the more sense it made to me."
Kirsch claims that his solution will only get more effective as more users come online using it.

Might make them ripe for a takeoiver, giving him the time to solve that cancer problem.
Back to top Go down
View user profile
Conrad



Number of posts: 5647
Location: Amsterdam, the Netherlands
Registration date: 2007-07-21

PostSubject: Re: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 8:31 am

'Kirsch, who is suffering from a rare form of blood cancer (another problem he is intending to solve),'

great sentence. Julian Simon would be porud!

dont know who his anti-spam-system works though
Back to top Go down
View user profile http://koenswinkels.weebly.com/index.html
Guest
Guest



PostSubject: Re: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 8:37 am

Conrad wrote:
'Kirsch, who is suffering from a rare form of blood cancer (another problem he is intending to solve),'

great sentence. Julian Simon would be porud!


Indeed!

- NonE
Back to top Go down
Zebra Foal



Number of posts: 899
Registration date: 2007-08-16

PostSubject: Re: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 12:31 pm

Conrad wrote:
'Kirsch, who is suffering from a rare form of blood cancer (another problem he is intending to solve),'

great sentence. Julian Simon would be porud!

dont know who his anti-spam-system works though


OK, well Kirsch has deployed much of his considerable wealth to good causes (through a Foundation named, I believe, after him and his wife.) He subsequently was diagnosed with a very rare kind of cancer--so rare that pharmaceutical companies have no interest in working on it. So now he has diverted his wealth to research into his own condition and its amelioration. What's wrong with that?

Re my question: can anyone explain Kirsch's method-- which is based on recipient analysis, not on sender ID (for spam filtering, not cancer!)? I really need to know.
Galets? Mike? Dylboz?

not a joke.... (rewards are possible....
Back to top Go down
View user profile
mike barskey



Number of posts: 1399
Location: CA
Registration date: 2007-09-07

PostSubject: Re: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 12:36 pm

I could not, tell from that article and one other, anything about his technology. I'm more of a power user than a true computer geek, but the 2 phrases he uttered about the technology itself was too vague for me to even get an idea how his spam filter works. That might be his goal, though - trade secrets and what not.

Quote:
We were sitting around thinking of ways to obfuscate the description about how our system worked so the spammers would be misdirected.


If he is trying to use obfuscation and misdirection to protect his methods (which won't work in the long run - someone will reverse engineer it), he won't want to announce them in an interview or article.
Back to top Go down
View user profile http://mike.barskey.net
Zebra Foal



Number of posts: 899
Registration date: 2007-08-16

PostSubject: Re: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 1:42 pm

mike barskey wrote:
I could not, tell from that article and one other, anything about his technology. I'm more of a power user than a true computer geek, but the 2 phrases he uttered about the technology itself was too vague for me to even get an idea how his spam filter works. That might be his goal, though - trade secrets and what not.

Quote:
We were sitting around thinking of ways to obfuscate the description about how our system worked so the spammers would be misdirected.


If he is trying to use obfuscation and misdirection to protect his methods (which won't work in the long run - someone will reverse engineer it), he won't want to announce them in an interview or article.


Thanks, Mike. This is helpful. Name your reward!
Back to top Go down
View user profile
galets



Number of posts: 213
Registration date: 2007-10-22

PostSubject: Re: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 2:18 pm

I haven't heard of such method, and, more: I would doubt it's efficient. What you're looking at spam arena is an AI vs human brain. And it's a losing fight too. I've been working as e-mail administrator for many years, spam was never getting even close to defeated.

Currently the e-mail is routed using SMTP protocol, in which anybody can email anybody. Emails are relayed using same protocol, which means that it's not even possible to be certain of origin of the email. But this is not even the root of the problem. I think people just don't care. It's the same situation as with government. As long as they are only mildly irritated with spam, nobody will do anything about it.

Contrary to the common beleif, it's possible to eliminate pretty much 100% spam right now, today. One of technologies which can ensure that is a VERY old one, it's called S/MIME, also known as "digital signature". Pretty much every email client supports that. All you need to do is obtain digital certificate for email signing and start using it. Encourage others to do same thing, then start rejecting all unsigned emails. Set an auto-response "this email address only receives signed messages". Easy as that. We have yet to see a spammer who will bother obtaining a certificate, and if he does, he's trackable.

But noone cares...
Back to top Go down
View user profile
Zebra Foal



Number of posts: 899
Registration date: 2007-08-16

PostSubject: Re: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 2:37 pm

galets wrote:
I haven't heard of such method, and, more: I would doubt it's efficient. What you're looking at spam arena is an AI vs human brain. And it's a losing fight too. I've been working as e-mail administrator for many years, spam was never getting even close to defeated.

Currently the e-mail is routed using SMTP protocol, in which anybody can email anybody. Emails are relayed using same protocol, which means that it's not even possible to be certain of origin of the email. But this is not even the root of the problem. I think people just don't care. It's the same situation as with government. As long as they are only mildly irritated with spam, nobody will do anything about it.

Contrary to the common beleif, it's possible to eliminate pretty much 100% spam right now, today. One of technologies which can ensure that is a VERY old one, it's called S/MIME, also known as "digital signature". Pretty much every email client supports that. All you need to do is obtain digital certificate for email signing and start using it. Encourage others to do same thing, then start rejecting all unsigned emails. Set an auto-response "this email address only receives signed messages". Easy as that. We have yet to see a spammer who will bother obtaining a certificate, and if he does, he's trackable.

But noone cares...


Thanks, galets! What about platforms that require the end user to undertake complaine methids to optimize the platform's efficacy?

Any ideas on why end-users don't contribute to the feedback loop that would at least reduce the spam plague?--ie. human factors?
Back to top Go down
View user profile
mike barskey



Number of posts: 1399
Location: CA
Registration date: 2007-09-07

PostSubject: Re: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 2:56 pm

I know what digital signatures are, but I didn't realize they could be used to prevent spam (as long as all my friends emailed me only using digitally signed emails). It's similar to white-lists. White-lists are lists of email addresses that you pre-approve to accept email from. This requires a special email client or a service (like SpamArrest.com), because whitelists are not built into the SMTP standard. When someone sends an email, the special email program or service receives it (not you) and forwards it to you only if the sender is on your list. If they are not on your list, their email is put on hold while the sender is automatically sent an email with instructions to add themselves to your list (if you allow that). This way, only humans can send spam, which greatly reduces the total amount of spam.

But all of this likely has nothing to do with Kirsch's method.
Back to top Go down
View user profile http://mike.barskey.net
Zebra Foal



Number of posts: 899
Registration date: 2007-08-16

PostSubject: Re: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 4:44 pm

mike barskey wrote:
I know what digital signatures are, but I didn't realize they could be used to prevent spam (as long as all my friends emailed me only using digitally signed emails). It's similar to white-lists. White-lists are lists of email addresses that you pre-approve to accept email from. This requires a special email client or a service (like SpamArrest.com), because whitelists are not built into the SMTP standard. When someone sends an email, the special email program or service receives it (not you) and forwards it to you only if the sender is on your list. If they are not on your list, their email is put on hold while the sender is automatically sent an email with instructions to add themselves to your list (if you allow that). This way, only humans can send spam, which greatly reduces the total amount of spam.

But all of this likely has nothing to do with Kirsch's method.


Thanks, Mike-- thought-provoking and valuable. There appear to be many instances in which the end-user does not comply with protocols associated with a particular platform to reduce spam. Do
you have any experience of that in your work? If so, do you have ideas on why end-users do not comply when--at least theoretically-- it is in their best interests?
Back to top Go down
View user profile
wilheldp



Number of posts: 191
Registration date: 2007-10-12

PostSubject: Re: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 5:23 pm

I have no idea if this is where he is coming from, but here's my crack at it.

As an avid email user, I get a ton of email every day. I could classify the email I get in to 3 categories...

1. Emails from friends, colleagues, and acquaintances. The content of these emails don't really matter, and vary wildly, yet I will open every one of them, usually immediately upon receipt.

2. Emails from memberships, forums, stores, feeds, etc. to which I belong, and have either voluntarily or involuntarily signed up for some sort of email notification. Depending on my mood, the sender, and expected content of the email, I might or might not open it, and even if I do, it may not be soon after I receive it.

3. Spam. Never open it...delete it as soon as I spot it.

Using information such as frequency of opening and how long it takes me to open emails with certain subject matters or senders, a filter could build a profile based on my email preferences. It could then use this profile to automatically segregate my mail to different folders. Combining that profile with other common Spam identification algorithms (key words, know senders, etc.), I think it would make for a fairly effective Spam filter.

_________________
I philosophize, therefore I evangelize
Back to top Go down
View user profile
mike barskey



Number of posts: 1399
Location: CA
Registration date: 2007-09-07

PostSubject: Re: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 6:05 pm

I think most people don't actively participate (or don't participate better) in anti-spam measures because they don't care too much and because it's not easy enough. Digital signatures currently require the use to know what they're doing, both in setting one up and in receiving digitally signed emails. It's not tough, but most people who use emails are not computer geeks, or even power users. Whitelists are also not easy; you have to subscribe to a service (costs $), in which case the technical work is done for you but you still must enter the emails you want to whitelist, you must maintain your list as you add new friends, etc.

It may be similar to many people's view about government: many people know or don't care that government is evil and makes things worse, but they're comfortable enough in their lives so they don't want to do anything about it - too much effort.

FYI: I have email habits very similar to wilheldp's.
Back to top Go down
View user profile http://mike.barskey.net
galets



Number of posts: 213
Registration date: 2007-10-22

PostSubject: Re: Anti-spam invention: How DOES it Work?   Thu Dec 13, 2007 7:39 pm

Zebra Foal wrote:
What about platforms that require the end user to undertake complaine methids to optimize the platform's efficacy?

Any ideas on why end-users don't contribute to the feedback loop that would at least reduce the spam plague?--ie. human factors?


Not sure what you meant by question #1?.. All mainstream email clients (Outlook, Outlook Express, Thunderbird, Eudora, name it) support S/MIME. All you need is digital certificate, which you can a) generate yourself (self-signed) or b) have some other internet authority sign.

Regarding human factors - there's already enough said in posts above.

mike barskey wrote:
I know what digital signatures are, but I didn't realize they could be used to prevent spam (as long as all my friends emailed me only using digitally signed emails). It's similar to white-lists.


Naw.. Not like whitelists at all. There are some major differences. Current black/white list services are based on the host/network(s), not "from" addresses, because you cannot guarantee if a guy in "from" fiels is the one who wrote email. But you can pretty much guarantee authenticity of signature. it is therefore possible to reasonably provide a service that tracks confirmed spammers, or their signatures. Current way of tracking spam leaves a lot of hosts a target for malicious false positive. Another implication is: networks who won't mind to shelter anarchists against the will of bullies, would most probably also shelter spammers. Digital signing, therefore is an only method so far to automatically distinguish one from another. Obtaining digital certificate from respectable provider requires a minimal, but manual effort, which makes spammer actually WORK, and make him trackable, which they don't like doing. But the same digital certificate can clear an email coming from "shady" segment, because now there's a much stronger indication of authenticity than an originating network. It's virtually impossible to fake digitally signed email.

... oh, and have I forgotten to mention this technology can be used for encryption of email correspondence too?
Back to top Go down
View user profile
Zebra Foal



Number of posts: 899
Registration date: 2007-08-16

PostSubject: Re: Anti-spam invention: How DOES it Work?   Fri Dec 14, 2007 2:43 am

wilheldp wrote:
I have no idea if this is where he is coming from, but here's my crack at it.

As an avid email user, I get a ton of email every day. I could classify the email I get in to 3 categories...

1. Emails from friends, colleagues, and acquaintances. The content of these emails don't really matter, and vary wildly, yet I will open every one of them, usually immediately upon receipt.

2. Emails from memberships, forums, stores, feeds, etc. to which I belong, and have either voluntarily or involuntarily signed up for some sort of email notification. Depending on my mood, the sender, and expected content of the email, I might or might not open it, and even if I do, it may not be soon after I receive it.

3. Spam. Never open it...delete it as soon as I spot it.

Using information such as frequency of opening and how long it takes me to open emails with certain subject matters or senders, a filter could build a profile based on my email preferences. It could then use this profile to automatically segregate my mail to different folders. Combining that profile with other common Spam identification algorithms (key words, know senders, etc.), I think it would make for a fairly effective Spam filter.


Thanks, Wilhelpd. This is interesting and useful re Hirsch's idea.
I have another couple of questions. If you care to consider I will be very grateful---. In some current spam filter platforms the spam receiver --the very end recipient, not the company's traps and filters, say-- *could* do some things to optimize reduction of how much spam their inbox shows and that they must delete. (Without much knowledge), I'm thinking of bayesian feedback-- a bit the way a user of voice recognition software "trains" the software to adapt ever more precisely to his voice. (I don't know how this is done and I understand that the "training--precision progress" re a spam filter platform would be quite different.

Here's my question: why do people inundated with spam who *have* the technology to at least shut out a lot of it, not do the things (setting up key words, etc.) that build the barrier or the sorting out of spam from good? Is it lack of knowledge? Is it a fear of false positives; is it a form in the workplace of diversion, off-task, self-proprietariness; is it that they are overwhelmed?

Any thoughts on these would be most welcome.

Also--- what are the burningly interesting issues in IT today?--maybe especially from the perspective of human-machine interface and human behaviour?
Back to top Go down
View user profile
Guest
Guest



PostSubject: Re: Anti-spam invention: How DOES it Work?   Fri Dec 14, 2007 2:49 am

Zebra,

Are you aware of ITConversations.com ? They are a group of people recording and podcasting conferences and interviews and such stuff on the tech industry and some of the podcasts are by brilliant minds on the cutting edge of this and that. You might want to check them out. I get an RSS feed that lets me know what cool stuff I might wanna check out.

- NonE (being sorta on and sorta off topic, as usual...)
Back to top Go down
 

Anti-spam invention: How DOES it Work?

View previous topic View next topic Back to top 
Page 1 of 2Go to page : 1, 2  Next

 Similar topics

-
» [Anti SNSD] Nh?ng hình ?nh ?i sâu vào lòng AFS
» Cold work permit & Hot work permit
» spam did in the last forum for depressivepersonality.com
» How can we contact GPPB when our emails are returned as SPAM??
» riding to work

Permissions in this forum:You cannot reply to topics in this forum
Liberating Minds :: Intellectual :: Technology-